Add initial draft of ML overview document

[mrocklin]

See #11095

cc @jrbourbeau @scharlottej13

This isn't yet done. Some TODOs:

• Make better example for Optuna
• Make better example for XGBoost
• Add more example links throughout

But I figured I'd pause here and ask for large scale feedback

[mrocklin]

I also added a video, but maybe that's a distraction. Mostly I used it because I saw it in the Optuna docs

[github-actions]

Unit Test Results

See test report for an extended history of previous test failures. This is useful for diagnosing flaky tests.

     15 files  ± 0       15 suites  ±0   3h 24m 49s ⏱️ + 3m 40s
 13 123 tests + 2   12 191 ✅ + 1     931 💤 ±0  1 ❌ +1 
162 492 runs  +24  142 455 ✅ +14  20 036 💤 +9  1 ❌ +1 

For more details on these failures, see this check.

Results for commit 456d3bd. ± Comparison against base commit aabc53d.

♻️ This comment has been updated with latest results.

[mrocklin]

TODOs are done. I think that this is maybe ok. @scharlottej13 if you're around and can take a look that would be welcome. I'm hopeful that this will help with dask.org links.

[scharlottej13]

(could be for a future iteration) I could also see including a joblib + scikit-learn section. Maybe there are also utility functions from dask-ml that people commonly use?

[mrocklin]

(could be for a future iteration) I could also see including a joblib + scikit-learn section. Maybe there are also utility functions from dask-ml that people commonly use?

I don't really see this all that often (xgboost and pytorch seem to be more common with dask workloads). I could be wrong though. I have no objection to people adding stuff here in the future.

For now I'm going to merge this.